The purpose of this research is to determine whether the risk of transmitting Cytomegalovirus (CMV) infections to premature infants by means of blood transfusion can be decreased by the use of pre-screened donors who have been shown to have no measureable antibody to CMV both by complement fixation (CF) and indirect hemagglutination (IHA) techniques. The incidence of transfusion-acquired infection in premature infants receiving blood from ransom donors whose antibody status will be determined retrospectively. In addition, excretion of CMV antibodies as measured by a precipitin test as well as the presence or absence of antibodies as measured by the CF and IHA techniques will be evaluated as risk factors for transmission of infection. Infants acquiring CMV infection will be followed for symptoms complicating their underlying illness, particularly for evidence of pneumonia, hepatitis, marrow dysfunction and poor weight gain. Following discharge, household contacts of infected infants will be followed for serological evidence of transmission of CMV infection from the infant to susceptible contacts, particularly those in the childbearing age group.